Hacker News new | past | comments | ask | show | jobs | submit login
How the car industry hid the truth about diesel emissions (theguardian.com)
170 points by NeedMoreTea 5 months ago | hide | past | web | favorite | 151 comments

I remember Mazda having trouble making their diesel engines performant while conforming to California emissions regulations, and wondering why they were having so much trouble compared to European car manufacturers. Then this scandal broke and it was obvious why.

I wonder why mazda or some other competitor didn’t find this by doing their own research? I make software for structural engineering and when some competitor launches a program that makes “better designs” it has usually turned out to be because they simply don’t follow the design codes/regulations. We wouldn’t be pushed from the market by that - we’d fund the analysis, reverse engineering and legal processes necessary to show the competitor is cheating.

My conclusion, from the first time I thought about it, is that they were all cheating.

Engineers move job; they talk to each other. Every car company had to have people who knew the truth and, if they were being unfairly competed against, could publish this truth.

Correct. Every single manufacturer in the industry knows what the others are doing. They all buy each other's vehicles and tear them down to the last bolt and bit to see how they work and how they can "borrow" some ideas without infringing on any patent (or breaking the law maybe).

They were all aware of what VW was doing and it wasn't a surprise, lots of companies did it [0] a lot of times [1]. Usually though the penalties were light. Many times they involved US companies and the strong local lobby pushed for the lighter fines. Like the heavy duty diesel engines case where 7 truck manufacturers were ordered to pay $1bn together for doing the exact same thing. Which in the grand scheme of things was peanuts compared to VW's $4.5bn (plus the tens of $bn of extra costs). So the expectation was always "it's just the cost of doing business, pay it and let bygones be bygones".

[0] https://www.nytimes.com/1998/10/22/us/record-penalty-likely-...

[1] https://en.wikipedia.org/wiki/Diesel_emissions_scandal#Previ...

> My conclusion, from the first time I thought about it, is that they were all cheating.

Mazda did, in fact, admit to cheating in emissions testing.

In other words, yes, pretty much the entire automobile industry cheated on those tests to varying degrees. The current "iteration" was also by far not the only one.

Its even simpler- they send each other prototypes, way before release of a new model, so everyone can reverse engineer that.

If Mazda was attempting and failing to get their cars to pass the defined emissions test, one can imagine they were only ever testing their own and their competitor's cars under those test conditions—which we now know triggered the cheat mechanism in VW cars.

But yes, had the research group not performed their on-road tests, it would have only been a matter of time before an honest car maker would have realised their competitor's sums don't add up and perform or fund their own independent investigation.

(Perhaps this is exactly what happened behind the scenes? It's known that CARB ordered the research due to discrepancies in US v EU numbers—but someone pushed them to look at the numbers closely. They put a lot of work in to perform these tests so it couldn't have been on a whim.)

> It's known that CARB ordered the research due to discrepancies in US v EU numbers—but someone pushed them to look at the numbers closely.

This is revealed in these quotes from the article:

German and his colleagues at the International Council on Clean Transportation (ICCT) in the US just wanted to tie up the last loose ends in a big report, and thought the research would give them something positive to say about diesel. ... German’s group also forwarded the findings to the US Environmental Protection Agency (EPA) and California’s Air Resources Board (Carb). “We were definitely scared. We wanted EPA and Carb to take over.”

Were they having technical trouble or just having trouble finding a cost-efficient solution?

AFAIK the technical solution (Urea injection) works fine, but it's expensive and VW didn't want to include it in all cars.

VW uses urea injections, but they wanted to make it transparent to the end user. The urea was to be refilled during maintenance, and disconnecting the system was the only way to make the tank last that long.

Not all TDI that were part of the emissions mess use urea, they also have a system that injects diesel into the particulate trap and burns out all the particulates at high temperature. Look up DPF regen. Note: I owned one of these cars, this is firsthand knowledge.

European here. My next car will be electric, likely a Tesla. The dieselgate is the reason why I'm not even considering Volkswagen's upcoming 'ID' family of electric cars. I've similar feeling about the whole traditional auto industry.

My next car will be "no car at all". Even electric cars have huge externalities that you do not pay for, but your children, namely, the disposal of the batteries, pollution created by the manufacturing, costly extraction of rare earths etc. Instead we should heavily invest in high-speed trains.

Except lithium ion batteries are very recyclable, and rare earth elements aren't in limited supply, they're just usually very sparsely distributed.

Wear from car tires cause the majority of microplastics that get into the seas (around 50%, if I'm reading understanding this study correctly):


At least in 1st world countries. The contribution of those are rather small compared to China and India.

WorldBank Pollution data: https://data.worldbank.org/indicator/en.atm.pm25.mc.m3?end=2...

Maplecroft Deforestation Risk 2018 (as opposed to his 2012): https://www.maplecroft.com/insights/analysis/esg-deforestati...

Statista Countries Polluting Oceans (updated version of his): https://www.statista.com/chart/12211/the-countries-polluting...


Also, manufacturing and charging li-io batteries for cars currently still produces more CO2 than petrol vehicles:


The biggest problem with cars, at least in European cities is the amount of space they take. Despite rising rents, making some cities unaffordable for people working there, we still mandate parking spaces for every build flat and use large amounts of space for multi lane roads, tolerating emissions, noise etc. Electric cars don't solve most of that. Making it easier to live in cities without owning a car should be the focus, not electric cars

I think the industrial scale, cost efficient recycling bit is untried, and we don't know if it will come to wide use soon enough to help with co2 footprint of cars?

High-speed trains are wonderful where the population density is sufficiently high, but in more sparsely populated parts, the car (and, for longer distances, the plane) is likely the only viable solution for decades to come.

With that as a starting point, the cars and planes might as well be electric. (Or hydrogen powered!)

I think this reasoning pins down the wrong variables of the equation. You can pick your housing and location based on not owning a car (among other criteria). And it scales too. And for decision-makers it's responsible to nudge your citizenship to this direction.

>You can pick your housing and location based on not owning a car

Only if you want to live in a large city and don't do anything outdoors. I go skiing, hiking, mountain biking, camping 3-4 times a week. Where I live (the Alps), it is extremely difficult to not own a car. There is no infrastructure and the terrain is unfriendly.

There isn't a housing or location option in Europe that allows me to give up my car and have access to the things I like to do

Usually you can also comfortably live in a small city few km from its center and have good outdoor options.

People will need to compromise of course. The alternative, not beating global warming is a big compromise too.

> The alternative, not beating global warming is a big compromise too.

It's not either-or. Owning a private car in the Alps and using it for the occasional shopping or skiing trip is far from being the main source of CO2 emissions for their respective country, let alone for the world.

The cost structure of private cars (big fixed cost, thereafter low marginal cost per km) and convenience works against this. Yeah, there are always exceptions, but these planning decisions should be made by considering people in aggregate.

Re "not the main source" - the co2 emission pie is very fragmented. We can't afford to go after only the biggest source of emissions. This is the divide-and-not-conquer method of losing this battle, to consider all the little parts in isolation and on each decide that it's not significant!

> Re "not the main source" - the co2 emission pie is very fragmented. We can't afford to go after only the biggest source of emissions. This is the divide-and-not-conquer method of losing this battle, to consider all the little parts in isolation and on each decide that it's not significant!

You're right, we're all responsible, and we can't just point at the biggest offender and not do anything about ourselves. But at the other extreme of the spectrum, hyperbole and guilt tripping of people who could be wiped from the face of earth without making a statistically significant impact on CO2 emissions isn't right either. Simply framing things as "cars or the planet" is wrong on so many levels.

As per 2017 statistics, the eight Alpine countries (including Europe's biggest polluter, Germany) accounted for about 4% of the world's CO2 (a third of US's numbers), half of it due to Germany. I'm not going to look up the statistics but I believe a large proportion of Germany's output is due to industry, and aluminum production in particular. It does not help that their atomic energy fears have driven up coal power plants. Even so, all but one of the Alpine countries have reduced their CO2 output since the 90s. Meanwhile over the same time period China has more than quadrupled its CO2 emissions and is the top polluter, and CO2 per capita both in China and the US is higher than in Europe.

> Yeah, there are always exceptions, but these planning decisions should be made by considering people in aggregate.

But we still (thankfully) allow people to make individual choices. If that means we improve the infrastructure required for a car-free life and still allow those who live in the Alps to choose a car (while strictening the emission & efficiency regulation), it's OK.

Or is it? Then why cars specifically? Right now there's no such thing as a personal carbon budget, but there's a whole bunch of things that contribute to the pie. The status quo is that if you're rich, you can consume more (income vs consumption): http://www.stat.fi/tietotrendit/media/uploads/tt2018/nurmela...

So far we've been very much accepting that some people produce more emissions than others. Should that change? I think that is something we should address before we ask if one should be allowed to choose to live in the Alps in a place where getting by without a car isn't feasible. In any case, private transportation in that region is not killing the planet. And for that matter, motorsport isn't either. I don't think we need to plan around those things, there are significantly more pressing areas to consider.

I personally think online car rental is very adequate for this. Locally, there's a number of very affordable services (usually there's tiers of membership fees that gives you steep discounts) that do this and I know many people that don't own cars but instead use this service somewhat regularly. Especially the flexibility of being able to pick up any type of car at a whim is very useful.

Assuming one can get a house at an affordable rate to start with.

It doesn't scale because not everyone is rich enough to live close to transportation hubs nor they can take everyone that wants to live nearby.

Housing supply and transport infra mostly responds to demand without pricing out people. Outside of urban hotspots where rent seeking rules.

Most cities, and most people's home vicinities, are not land cramped like the world's megacities.

Come to Europe to the cities where one actually finds a job and then try to use the public transportation for that 1h 30m commute time, easily done in 30m with a car.

Already there, but it's not like that. 1h 30m in public transport would get me to another town.

If you are lucky enough to live in suburbs with a job that works for you and good connection, then all the best.

Just don't assume it actually scales for everyone, specially when the closest bus stop requires 15m walking for a bus that only comes around every 30m, just to get to the next connection point.

While urbanisation overall is probably a net plus for the environment, do keep in mind that, say, farming is quite area intensive.

Someone needs to work at those farms - and they are not going to commute to work by high-speed rail.

(Granted, there are experimental urban farms out there, but I don't see those feeding the planet on a short-term scale.)

Farming is area intensive, but the number of farmers we need has been steadily declining.

> And for decision-makers it's responsible to nudge your citizenship to this direction

Is the flip side of this "People who want to live in solitude are not qualified to make decisions"?

You can live in solitude quite ecologically without a car too of course. Nudge =! force

"Nudge" is the 21st century version of "force", as it's been realized that "force" provokes backlash

It means a different thing.

> You can pick your housing and location based on not owning a car

If all I ever wanted from life was to go to work and back to my apartment, via a grocery store.

As a matter of fact, I can't choose where my friends and relatives live, and I'd probably have to drop some of my hobbies too if I were to live without a car.

You will probably have to drop some of your hobbies too if we don't manage to limit Global Warming to below 2 degrees.

Yeah, well, I did things the other way around and picked my job around not having a commute. That means remote, I work from my apartment.

I still have a car, but I drive much less than those who commute by car every day. It might be sitting unused in the lot for weeks (I have my feet & bicycle for grocery store trips). But when I need it, I need it, the public infra and ridesharing simply isn't there. (But I occasionally give rides and haul stuff for people who don't own a car, hey, isn't that exactly what we need?)

As far as global warming is concerned.. well, one glance at the statistics shows that my car ownership couldn't matter less. For example, if you take the top polluters (China & USA) and compare to my country (Finland), and break it down by sector, you'll find that our transportation's contribution to CO2 emissions in the world is a fraction of nothing.

As far as domestic energy consumption goes, we've cold winters, and heating is the biggest drain. There's a lot that could be done to improve the energy efficiency of older homes without spending too much money but the will isn't there. The only way we get heat efficient buildings reliably is via regulation that applies to new buildings.

It doesn't make sense to quantize emissions by country to decide whether your choices matter. (Or by by other attributes)

Huh, why not?

Things people do generally become a problem only at scale. If the scale isn't there, there's no problem. There is no global law, everything we regulate is regional. The consequences of regulating a speck of noise in the statistics does not matter; regulating a major contributor does.

I can go on a camping trip and light an open fire in the woods, no problem. My neighbor can do so too, no problem. If all the 1.3 billion people in India started doing it, we'd probably have to do something about it.

This applies to pretty much everything you can do. In Californian drought, it may be occasionally necessary to restrict the use of water for watering lawns or washing cars, but here we don't give a shit because we aren't running out of clean water. In areas with expensive desalination, all water use is rather consequential.

There are low density regions where it's ok to heat your home by burning wood, it will never cause an air quality issue and the planet won't die as long as the world's population is not concentrated in these areas doing the same thing (then it would not be a low density region).

There are high density cities where they're starting to reroute traffic around it and collect tolls from those who insist on driving through, and there are low density areas where it doesn't matter.

There are areas where little to no energy is spent on heating homes, and then there are areas like Finland with cold winters and a lot of home heating.

The attributes are absolutely relevant, we need to put things into perspective and attack problems where the scale is an issue. Planet earth doesn't care that you're pure of heart, or that you breathe less CO2 than your neighbor, it cares about the absolute quantity of greenhouse gases that are shot into the atmosphere. That is not solved by fighting the fight where the quantity is relatively zero. And yes, that means the world is unfair, and you may be subject to more restrictions depending on where you happened to be born, or where you chose to move. Holding everyone on the planet to the same standard was never a thing (unfortunately the globalists don't seem to understand this and we're occasionally suffering the consequences of regulation made by & for people a thousand miles south of us, in a different climate, thanks EU).

The change is going to have to affect your friends and relatives too.

You can always adjust your hobbies, or work. Global warming is more important.

Doesn't mean they'll be living next door all of a sudden!

This is ideal.

Unfortunately the society has been pushed into “car as a requirement” territory for ages.

I commend the effort, and I’d like to not need a car... perhaps I don’t, but it will require a few major changes.

It’s crazy that car pools are not more common, btw.

> Instead we should heavily invest in high-speed trains.

This is ideal for obvious reasons. If only cities and governments (especially in the U.S.) would see the value. But alas it seems like only some countries can pull it off.

Also, most EV batteries today are highly recyclable and with fewer moving parts they can last much longer than ICE cars.

I love investment in trains but in the USA, small improvements to investment wouldn't be close to enough. We need EVs because the political will is simply not there for any infrastructure spending. Its either EVs or more emissions, sad state of affairs.

I drove a VW passat (bio/ngt gas hybrid) for years.

This stuff pissed me of so much that together with the Exxon stuff I will never own a VW or a fossil ICE car ever again.

I went for the Leaf as I can fit a double buggy in the trunk and it has a good enough range. Not cheap but it’s what I can do as a consumer in this liberal market economy.

Good job, greedy bastards! :)

Side note - using an EV have given me for the third time kind of an technology epiphany experience:

1. Internet

2. The first iphone

3. Now, the EV

Charging in the garage over night, never visting a gas station other than for coffe on the road. It’s awesome!

What’s so incredibly convenient is that the grid is built out already, i.e the infrastructure is in place, and has been for decades.

Imagine someone suggesting we haul this toxic and highly flammable stuff called “gasoline” around the world to “stations”... what a ludicrous idea!

I see parallells to music and media distribution and the internet.

This whole imbroglio was a painful learning experience for me. For several years before I bought my used jetta TDI, I started looking into Bio Diesel. I still have 12 Jatropha Curcas trees in my back yard, which were to be the start of a possible startup. I am greatful for the learning experience, but I keep waiting for the wisdom that is supposed to make up a little for the aggravation of aging.

Teslas are still huge in mass (2000+ kg) and space/dimensions compared to what's needed from a car, and have a correspondingly big carbon footprint from building the thing and batteries. We should aim for 500 kg cars. We could do a 4 person 500 kg car in 1957 after all (Fiat 500).

> We should aim for 500 kg cars.

So you want a Tesla without batteries?

A single battery cell weights approx. 200 lbs (90 kg). The Model 3 has four. The total weight of the batteries is thought to exceed 1000 lbs (inc. housing, harnesses) or about 500 kg.

Then add on all the legally mandated safety technology a vehicle from 1957 wouldn't have (front airbags, seat belts, windshield defroster/washer, taller seats, and much higher crash absorption/roll support).

You could absolutely build a 500 kg kit-car that has 1957 safety. But no commercial vendor can sell you a vehicle even in that ballpark.

There would be less need for safety technology if everyone was driving around at 60kph, not 100kph. A lighter battery would last much further.. lower cost vehicles.. less environmental impact.. less noise around populated areas etc. It would take longer to cover long distances, but.. smell the roses, or take a train? Pipe dream obviously.

> There would be less need for safety technology if everyone was driving around at 60kph, not 100kph.

That's like saying you don't need to wear a helmet while bicycling as long as you don't go faster than X.

When in reality just falling over while standing still, can already give you quite a good knock on your head, depending on where you are landing with it. Just like you can't control the speed of other traffic participants.

Even if you lowered the allowed speeds across the board, people would still end up speeding, probably even more so than they are doing already.

In that context safety features should never be "optional extras" [0] because they usually are not "trade-offs" but straight up improvements.

[0] https://www.nytimes.com/2019/03/21/business/boeing-safety-fe...

There is very effective technology to prevent speeding. We just need to build more of it.

Just take the Renault Twizy (a production 2 seater 500 kg ev on market since 2012) and try harder in the same direction. The regulations allow it.

To preempt some protests - You could put a trailer on it for occasionally transporting more stuff.

Model 3 weights 1600...1800 kg (SR/LR), not much more than similar spec'd ICE car.

My next car will be electric too, but I won't exclude VW entirely. They did the most obvious cheating, but basically all Diesel cars would use some schemes to pass the exhaust tests. Their ID line seems to be a genuine attempt at going full electric. I would of course triple-check any claims about their cars going forward.

Unfortunately you seem to be in the minority. People are still happily buying VW.

Why "unfortunately"? Would it be better if every company that does a bad thing was forced to just close shop?

If that be the case I'm pretty sure we'll have to shut down Tesla within a couple of years over their "Autopilot"/"Full Self Driving" claims.

"Do a bad thing" is a pretty major understatement of the degree of criminality we're talking about here. Over-representing the capabilities of Autopilot is a totally different thing that embedding a defeat device to defraud every national regulatory agency.

If you excluded every manufacturer that did this at some point, you're pretty much only left with Tesla and probably some other smaller new EV companies.

This speaks volumes.

... and I wouldn't be all that sure that none of the EV folks aren't cheating on their range tests ...

>Over-representing the capabilities of Autopilot is a totally different thing

Citation needed. They didn’t do that.

As convenient as it might be to burn VW at the stake for this as the "sacrificial lamb", that misses the point.

Because VW ain't the only ones who engage in these kinds of tactics, by now it seems like it's some kind of "open industry secret" how manufacturers try to get the best measuring results during environmental testing for those good ratings, which directly translates to good marketing.

> embedding a defeat device to defraud every national regulatory agency.

Cite this claim.

> Cite this claim.

It is, literally, a large portion of TFA.

You may have misread what Tesla says about these technologies. They state very clearly that FSD is not here yet, and they also are very clear about the limitations of Autopilot. The problem is that people (like you, for example) mistakenly assume there are “claims” beyond the statements that are made.

> They state very clearly

They state it in the fineprint. This is what the Tesla Autopilot page used to say for years (this is a 1+ year old screenshot) [0]. Which part of that is accurate? How clear are they about the limitations and the real capabilities of the system? Tesla took the narrow areas where those systems can do a reasonably good job and extrapolated to the whole self-driving experience, blaming regulations for the bits that aren't really there. No autonomous system in the world at this time is capable of anything close to what a human is.

They still claim that:

> Your Tesla will figure out the optimal route, navigate urban streets (even without lane markings), manage complex intersections with traffic lights, stop signs and roundabouts, and handle densely packed freeways with cars moving at high speed

Which is patently false.

[0] https://imgur.com/a/bsgxCCn

> No, they state it in the fineprint so to speak.

Self-Driving is not even released yet, prove me wrong otherwise. What you're attempting to quote are forward looking statements. Maybe you need to read the document you linked, slowly this time.

I don't have to prove anything, I know they're not flat out lying [0], they're just misleading. Many drivers (ask the dead) are/were under the impression that the car can drive itself. And Tesla doesn't do much to make this distinction clear.

[0] Except for the "we'll have it next year" bit. That's just to get a little more cashflow, pretty sure not even Musk expects to have FSD ready "next year" unless it's a perpetual "next year".

> I don’t have to prove anything

You can’t just make bold claims and not back them up... Makes you look imprudent.

> Many drivers (ask the dead) are/were under the impression that the car can drive itself.

Everytime you engage Autosteer. You get a warning that says “Always keep your hands on the steering wheel” and “Be prepared to take over at any time”.

More importantly, each driver assistance update also explicitly state the capabilities and you HAVE to toggle them on. No where in the car UX or manual even suggests that it can “drive itself”.

I think it’s you have the wrong impression. Unless you’re speaking anecdotally, which is I highly doubt.

> You can’t just make bold claims and not back them up

That's not how the burden of proof works. Tesla is claiming the following:

> All new Tesla cars have the hardware needed in the future for full self-driving in almost all circumstances.

> The future use of these features without supervision is dependent on achieving reliability far in excess of human drivers as demonstrated by billions of miles of experience, as well as regulatory approval, which may take longer in some jurisdictions.

Where is the proof of that? The suggestion is that everything is in place short of regulatory approval. And it's always "next year". The result is this [0]. People genuinely believing and insisting that car can drive itself. The result is the only driver fatalities while autonomously driving were in a Tesla.

Do you see how people interpret this as "well it's there but you know, pesky regulators want to delay it by testing and stuff". If the vast swaths of internet commenters and drivers that can no longer provide an opinion is not proof enough that the statements are misleading I don't know what is [0] (again). You may see through them but it doesn't make them less misleading. That whole page is designed to give an inflated impression of the state and capabilities of the system regardless of consequences.

Other manufacturers have at least equally capable hardware and software but none make these claims, because they are misleading and will cause deaths.

What would you think if someone sold you a mobile phone claiming it's able to crack any encryption no matter how strong? Of course it might take longer than the lifetime of the universe but hey, it is possible. Not lying, just misleading. So wanna buy a phone?

[0] https://www.youtube.com/results?search_query=tesla+self+driv...

> People genuinely believing and insisting that car can drive itself. The result is the only driver fatalities while autonomously driving were in a Tesla.

This is hilariously false. Do you really want to prove this? Go on the teslamotorsclub.com forums or teslamotors subreddit and ask actual owners if they think the car can “drive itself”.

Link the post and I will PayPal you $1,000 if you’re right. Because at this point you’re just going around in circles.

It’s hard to discuss these things when both sides don’t have all the facts.

Just as a basic background for what happens, please try this: sit in a recent Tesla, engage autopilot, and read what it says on the screen. Also try ignoring the advice the screen gives you (pretend you didn’t read the “fine” print as you call it, and see what happens when you try to let the car drive with no input.)

Do this, then we can talk about this stuff, OK?

> It’s hard to discuss these things when both sides don’t have all the facts. Just as a basic background for what happens, please try this: sit in a recent Tesla, engage autopilot, and read what it says on the screen.

My point exactly. Yet he seem to have pretty stong and absolute (albeit skewed) opinions. A true armchair quarterback.

Unless the dashboard tells me why the Tesla website and marketing (not to mention its CEO) make all the false/misleading claims I don’t think we need to talk about anything else.

xedeon 5 months ago [flagged]

> Unless the dashboard tells me why the Tesla website and marketing (not to mention its CEO) make all the false/misleading claims I don’t think we need to talk about anything else.

Honest question. Do you have a short position on the stock?


Were they? As I said (I'll say again it but slower) Tesla is the only manufacturer with multiple driver fatalities while "self driving". Plus the countless crashes with no fatalities.

Either the description of the Autopilot's capabilities is misleading making drivers assume it's more capable, or drivers with severe comprehension issues are over represented among Tesla drivers, to the point where they actually kill themselves. The car or the driver have a problem. You pick.

> Were they?

Yes, many times.

> Tesla is the only manufacturer with multiple driver fatalities while "self driving". Plus the countless crashes with no fatalities.

Again it's pretty obvious that you have no idea what you're talking about here. That statement is not only grossly false but also ignorant (seem like a theme with your comments). For one, autopilot features are really just glorified Lane Keeping Assist System (LKAS) and Traffic-Aware Cruise Control (TACC).

They are one for the umpteenth time NOT "self-driving" any Tesla owner knows this (prove me wrong otherwise my $1K offer still stands). There are ZERO indications that even remotely suggests that the car is "self-driving.

From the moment you take delivery and they show you the features of the car, to turning on each individual driver assist features and actually engaging TACC and LKAS. Zero, nada, none. So stop saying this. Go find any Tesla owner discussion group/forum on the internet and ask actual owners. You will quickly see how wrong your assumptions are.

More importantly, TKAS and LKAS have been around for a while now and the only significant difference with Tesla's version is the auto lane change functionality.

Below are just a few other safety systems that have TKAS and LKAS:

1. Honda Sensing 2. Toyota Safety Sense™ (TSS) 3. AcuraWatch 3. Some Mercedes Benz vehicles with the Driver Assistance Package

There have been plenty of fatalities with these systems as well. But guess what? you don't see them in the news.

Some Tesla owners on reddit even talked about the differences with these systems from their anecdotal experiences and found that Tesla's implementation was better (these are actual owners not marketing BS). i.e. Some won't even give any visual/audible warnings when disengaging.


No, you are just proving my point.

The screenshot is talking about future functionality that is not released yet, but you are mistakenly evaluating it as patently false when you have absolutely no way to know, since it isn’t released yet.

“System B will do X” is not invalidated by observing that System A does not currently do X.

> but you are mistakenly evaluating it as patently false when you have absolutely no way to know, since it isn’t released yet.

What makes it patently false is that Tesla hasn't proved it even after claiming "next year" for several years now. Even worse, they still don't and can't know if the system will ever be able to perform as described or if the hardware is enough as claimed. If the party making a claim doesn't have the proof for themselves then how can the statement be true today, when they claim it? They are selling today's assumption as a certainty. It doesn't say "we hope" or "be believe" or "it might". They say "it will" with no reservation and no proof available anywhere. That is the falsehood.

And just as a sidenote, the company that was providing Tesla with the hardware and knowhow when the statements were originally made, Mobileye, ended the partnership in 2016 because they did not believe their own systems are able to fulfill the promise. That shows that promises were already made once without Tesla having any proof.

Any other company doing this and people would be all over them with accusations. But Tesla has great PR, it is actually working for a better future, and people love a good cause enough to disregard the dirt under the rug.

This is all a bunch of fake outrage. Words are written. Some people skip reading important parts of those words, and then blame Tesla for their own laziness or poor reading skills. This is basically what you’re doing, although maybe more as an apologist for others who don’t care to read or pay attention.

I’m not even talking about fine print. The stuff that covers this is all very straightforward and not fine print at all. People should take responsibility and read what is actually said.

The outrage should be directed at people who make wild misinterpretations of perfectly reasonable statements.

Yes... it perfectly reasonable to make claims that even the supplier of your HW refutes. Really says a lot about your willingness to say whatever it takes to get more PR and push a few more sales.

In the meantime Tesla is the only company with autonomous driving driver fatalities. Multiple. Unless you’re going to claim Tesla buyers are more predisposed to... laziness and inability to process information, then something is definitely wrong in the way the system is presented. Something other manufacturers and all their buyers get right. Mostly all others insist overpromising and underdelivering can be dangerous.

They are not autonomous driving fatalities. They can’t be, because there is simply no autonomous driving. You are very confused. Autopilot is not autonomous driving.

It's self driving. It's not L5 but the car was taking the decisions in that limited set of conditions.

And that's the whole point, the car is unable to take the correct decisions but people still trust it to do so. Why would they? Unless you'll tell me Tesla drivers are intellectually challenged then it must be that they somehow get the wrong impression of what the car can do.

How come this doesn't happen to any other manufacturer building cars with this type of limited automation? I see a some people victoriously claiming I've been proven wrong yet all avoid answering this direct question. If nobody can explain why so many Tesla drivers are suicidal then the blame must lie with the system or the way it's described.

>If nobody can explain why so many Tesla drivers are suicidal then the blame must lie with the system or the way it's described.

I see. You are looking to explain the facts and you have three theories here. 1) Suicidal drivers; 2) Bad system; 3) Bad system description.

Let’s just agree it’s not #1. Effectively suicidal you could say, but let’s agree they were (likely) not actually intending suicide.

Number 2 is ruled out because the system was never designed to be fully unmonitored... blaming it would be like blaming a seatbelt for an injury sustained by someone who wears the seatbelt incorrectly.

Number 3 seems like a likely possibility on the face of it, but the system is actually pretty well described in practice so I would suggest another, fourth, possibility which does not appear on your list:

4) The drivers were imperfect humans who, as some will do, failed to heed the ample safety warnings and were unlucky enough to encounter situations where this failure led to disastrous results.

I’ll note this kind of thing happens all the time in cars without autopilot as well. You can point out that in those cases an autopilot system was not in the picture, but that’s a circular argument... of course it wasn’t. People will sometimes be foolish. No system can protect all fools all the time. Expecting otherwise is unreasonable. You’re just blaming Tesla because its autopilot is the first / most prominent such system that has been rolled out to production vehicles. Of course there will be a few foolish people amongst the hundreds of thousands of Tesla drivers. That’s on the people, not the system or the description. Nobody ever promised autopilot would be a perfect nanny that can fend off all dangers and control all human behaviors.

>nobody can explain why so many Tesla drivers are suicidal

Not suicidal. Just foolish. And not “so many.” We are talking about a handful of incidents, in fact fewer than can be counted on one hand.

Fake outrage.

I don't care about whether things are "ecologically sound" or not.

I know the earth is important and blah blah, but I have bigger fish to fry every day. I'm sure most people feel like that.


I work in this industry, so I've been following it from the start. I've run test cells and written software to analyze the results from the emissions cycles that the government uses to validate the emissions compliance. I have worked around and for a very large diesel engine manufacturer who I am convinced does not cheat on this test. You'll understand why, below.

I recently leveled-up my career, and I now work with the person who originally setup such testing at this manufacturer, and one of the two people who literally wrote the spec, which is one of the most thorough technical documents I've ever laid eyes on. These are 2 of the highest-caliber engineers I have ever met in my 25-year professional career, both in terms of intelligence and capability, and in terms of integrity and trustworthiness.

This quote from the article pisses me off: "What I understand now is that the people we entrusted with the power to protect us essentially decided not to bother. Instead, they have allowed carmakers to spew whatever they want into our air." Really?! The government commissioned the spec. The government mandated that engine makers make their engines comply, and validate this compliance with internal testing. The government then paid independent testers to validate the manufacturer's results.

Excuse me, but what more did the author expect the government to do? The regulators entrusted this process to expert engineers, and expert testers. Did the author expect people who could invent the spec and create a test cell to do the validation to go to work for the EPA directly, and do the paperwork, instead of the engineering? Did they expect them to escrow the code, and validate that there was no cheating? Possible, but, again, you're asking a very specific skillset to give up DOING, and instead REGULATING, and that doesn't usually work.

The fault of this scandal lies -- of course -- with the manufacturers, but the fault also lies with the legislators. To control diesel engine emissions within the spec requires an array of very expensive after-treatment components. That can work in the class-8 heavy trucking industry, but you can't even fit it into a passenger car, let alone make it affordable at gasoline car prices. To think that cars could use diesel engines, be cost-competitive with gasoline-fueled alternatives, AND pass the ridiculously-low emissions standards that lawmakers have arbitrarily decreed was simply wishful thinking.

>To think that cars could use diesel engines, be cost-competitive with gasoline-fueled alternatives, AND pass the ridiculously-low emissions standards that lawmakers have arbitrarily decreed was simply wishful thinking.

The emissions standards are perfectly sensible for maintaining good air quality. It definitely looks like it is wishful thinking for diesel engines to meet those standards and be cost-competitive with gasoline-fueled alternatives, but there is nothing wrong with that. Gasoline engines may simply be the more suitable technology for passenger cars.

> It definitely looks like it is wishful thinking for diesel engines to meet those standards and be cost-competitive with gasoline-fueled alternatives

The technology that would have brought "Dieselgate" diesels into compliance with EU emissions rules was said to cost manufacturers something on the order of US$100 per vehicle: Selective catalytic reduction (SCR) with a refillable AdBlue (urea) tank.

However, prior to Dieselgate there is alleged to have been collusion between the major German automakers and a secret agreement not to install it on small vehicles.

This collusion, rather than the emissions cheating itself, is the focus of ongoing criminal investigations in Germany. (Cheating on emissions tests itself was not actually illegal in Europe, because the legislation was created with loopholes).

As of 2019, AdBlue is becoming common on new small diesel vehicles in Europe.

The Euro6 standard was launched back in 2014, following the Euro4- and 5 standards that most of the diesel-gate cars adhered to. Most cars manufactured from around 2016/17 are Euro6, and as far as I’m aware Urea tanks are in use on most or all.

I honestly don’t understand why it would have been an issue to have it before. My diesel tank is 65L and it takes me over 1000km. Even if they had made a 10L Urea tank and shrunk the gas tank to make room for it, the range is still more than enough. Filling it takes 5 minutes once every 1-2 years and costs about the same as windshield wiper fluid. Can’t see why that would make the car unappealing either.

> Filling it takes 5 minutes once every 1-2 years

Isn‘t this a part of the problem? The urea tanks are way too small and need refilling way too seldom which means a too low urea usage? Volkswagen wanted urea to be refilled by your service station and not by the customers themselves because they might have found this inconvenient.

The problem wasn't just that it was an inconvenience. Under the EPA rules of the time, urea refills were considered critical emission-related maintenance. Initially, this meant that they had to last 100,000-150,000 miles unless the EPA gave a waiver allowing more frequent servicing intervals.[1] This wasn't physically feasible. A couple of years into Dieselgate, they decided they'd allow DEF refills at the oil change interval, which still required much larger tanks than the European models (typically replacing the spare tyre) or cheating.[2] It wasn't until mid-2014 that they decided users could be allowed to refill their own urea tanks and lowered the minimum DEF refill interval for cars to a slightly more reasonable 4,000 miles.[3]

[1] https://iaspub.epa.gov/otaqpub/display_file.jsp?docid=16677&...

[2] https://www.govinfo.gov/content/pkg/FR-2009-11-09/pdf/E9-269...

[3] https://www.govinfo.gov/content/pkg/FR-2014-08-08/pdf/2014-1...

My last car was a diesel-gate car (09 Audi) which didn’t have a Urea tank at all.

I know VW managers first thought Urea was an inconvenience to have at all, then an inconvenience for owners to refill so should last a service interval. My new car has a Urea tank and isn’t a diesel-gate car. It need to refill maybe 1 time before the service intervals. I can’t see why it would be considered inconvenient - it’s easier to refill than windshield wiper fluid (that you need to pop the hood for!). So if it would have come at a price of 20L diesel capacity space, and require refills every month, I still wouldn’t have found it inconvenient.

I don’t know how they determine how much AdBlue to consume now, but I’m guessing since they now manage to comply with regulation they just use as little as possible? If they can further limit emissions by using more, then there’s some headroom for improving too.

Exactly. If it's only being filled with (eg) 8L every 1-2 years, then it's not being used much and is probably not very effective at reducing NOx emissions.

Depending on the engine, driving style, etc, it should really be using something on the order of 20-60 litres of AdBlue per 1000 litres of diesel.

There were some pretty big licensing hurdles to getting the EPA to accept urea tanks as part of an emissions control system: https://iaspub.epa.gov/otaqpub/display_file.jsp?docid=16677&...

Also, note that the 2007 Tier 2 NOx emissions standards mentioned in that document are pretty much as strict as the 2014 Euro6 standard, at least for manufacturers which mainly produce diesel cars. The EU emissions standards were much weaker back then, giving manufacturers more time to develop tech to meet the stricter standard.

"Excuse me, but what more did the author expect the government to do?"

It is not some outrageous expectation that the $8,000,000,000+/year, 14,000+ employee EPA and/or DOT independently verify the real world emissions performance of passenger cars and other vehicles. The cars were so grossly out of compliance that anyone with a ordinary field emissions testing machine could have caught it.

If, as is very likely the case, the regulators had no technical resources among their hordes of lawyers they could have asked any number of publicly funded universities to perform the actual investigation and go beyond the agreed upon and easily fooled test regime. The discovery was actually made by US academics with very limited financing.

This institutional pencil whipping is SOP at EPA. GOA had a gasoline powered alarm clock (among other improbable devices) Energy Star certified as part of an investigation in 2010. Energy Star certifications are literally pencil whipped through an automated system with no human review.

When the EPA found itself answering for their part in the emissions scandal in public hearings there was no defense. They didn't even try to cite limited budgets[1] -- the go-to excuse every time some regulator comes up short -- as a cause. They just failed.

That's difficult to accept for some, and so all manner of rationalizations are employed to dilute that simple truth. But I'm not among them.

[1] https://www.c-span.org/video/?c4787386/blaming-budget

I still shake my head at this. If you wrote laws that said manufacturers of a particular thing had to do X, and expected them to adhere to it internally, and then paid expert testing facilities to validate those results, why would you EXPECT that you would have to do additional testing to validate the validation?

And further more, why would you think that merely testing the test would lead to a different result? You can say that the Euro V is "not representative" (as someone else did in this thread) of normal driving, but the FTP test's torque and RPM curves were lifted from real-world driving through hilly areas. It most definitely is representative. I don't know about the Euro V.

So I don't get it. In this scenario, what people are asking the EPA to have done was load up vehicles with mobile emissions equipment (which is a very difficult thing to do, by the way), and validate that the TEST ITSELF correlated with real-world driving. Granted, this is an interesting exercise, but one that I would expect the EPA to - again - pay someone else to do.

Given the engineering that went into the spec on the US side, I wouldn't expect that this was something they should have needed to do. You can say that they were the regulatory body in charge of this, and therefore should be responsible for catching cheaters, but I don't see how a reasonable person can find that the EPA was intentionally negligent in this matter.

Nobody charged the EPA with ensuring adherence to some made up test cycle, but to the emission limits. If the test cycle is your approximation of real world behavior, it is highly negligent to never confirm that real world results actually match the approximation! This is science 101.

I’m specifically arguing that the FTP cycle result IS, in fact, a real world result. So to me, claiming that the EPA needed to do their own verification would be silly. The only reason to do this would be to catch cheaters.

I know I’m naive, and very trusting. I would never have thought a company to have the utter stones to completely fake the results as VW did. If you’ve looked at the depths of their algorithms to do it, you see the effort that went into it. I just would never expect such behavior from one of the largest auto makers. It’s so far beyond the pale to my expectations that I can’t blame the EPA for not catching it.


I'm not a test engineer but work closely with them- two of their core ongoing job functions are monitoring real world performance to find test escapes, and adapting the tests to more closely drive real world performance.

"what people are asking the EPA to have done was load up vehicles with mobile emissions equipment (which is a very difficult thing to do, by the way):

No, that's incorrect in several ways.

There was/is no need for mobile equipment; simply deviating from the test regime on a stationary emissions test machine was sufficient to expose the defeat device. The defeat device worked by detecting the test conditions and defeating that detection is easy. Turning the steering wheel while testing, for example, will inhibit the VW defeat device. Simply running the test long enough (at least 26 minutes) would have exposed it with some common models.

Also, mobile testing is not 'a very difficult thing to do.'

For a time in the early 90's I was certified to perform emissions tests. At that time the actual machine fit on a wheeled cart; two men could hoist it into the rear of light pickup in a few minutes. I actually did that when we needed to move one to another site to replace another that had failed. That wasn't strictly legal (the units were certified for operation at a designated site) but the station owner didn't give a damn; he needed a temporary replacement unit and was not above fudging the paper work.

So nigh on 30 years ago any given wrench monkey with a power inverter could have operated an exhaust analyzer from a moving vehicle. Today the units are even smaller; now we have HAND HELD exhaust analyzers[1].

We've also had remote sensing emission testing equipment available for at least 10 years now. Output is detected as the vehicle drives by the unit; no need for anything on board at all.

Face it. There is simply no excuse. The regulator failed utterly. The regulator publicly admitted their failure without qualification. Your faith is misplaced and your rationalizations do not work. This is, assuming intellectual honesty, the point where you adapt your worldview, because the one you have is simply wrong.

[1] https://thecarlsoncompany.com/products/infrared-industries-h...

Didn’t know about those devices. Doubt they qualify for certification compliance, but I guess it would get you on paper.

We need better models of governance.

One party arrangements (self-policing) ain't working.

Two party arrangements (regulation) is hit or miss.

Maybe add a third chair to each negotiation. Just like the US Constitution has three branches of government. What would that look like?

Maybe incentivize competitors to narc on each other. What would it have taken for Mazda (for example) to blow the whistle?

1. massively weaken European defamation laws

2. make NDAs and noncompetes illegal

3. remove all taxes on journalists and jouranlistic organizations

4. force companies to put whistleblowers on their boards of directors.

Why do you think escalating emissions standards exist? The entire purpose is to gradually retire whole technologies!

It is essentially deprecation. The ultimate goal is always reduce to zero, you will find very few emissions in our continuous world that are somehow safe at arbitrary cutoff X but unsafe above.

I guess it depends a little on whether you feel like the spec writer or testers should have anticipated cheating, and attempted to detect it or not. If the job of a tester includes catching cheaters, or if a good spec should be very hard to cheat, I'm not sure missing that falls at the feet of the legislators.

How can VW create an engine, cheat, and sell it to Chrysler? Then Chrysler has to put a urea tank to meet emissions. How did some brilliant engineer writing the spec or the third parties not notice?

An SCR (and EGR, and DOC, et. al.) is a "given" in this equation. There's no hope of passing emissions without it. Even on gas engines, you still need a catalytic converter to pass.

My point is VW sold a TDI engine in the US. The same engine was sold to Jeep and used in the Grand Cherokee but with extra emission controls like urea. If I worked at Jeep or one of the 3rd party testers I would’ve made a lot of noise about VWs lack of a urea tank.

I live in france, and I often spot cars that have a grey/blue smokes. Many other cars also have this awful smell.

Here, every 2 years, cars must legally be examined in a "controle technique", like an exam which controls several aspects of a vehicle.

I wonder if those fumes are legal, and I really wonder if people can manage to cheat at that exam, because I think I'm noticing an epidemic of those fumes.

How can I really be expected to ride a bike next to cars like this? I'd rather spend time in a tram or bus than expose my lungs to this.

I have an older diesel VW. It is extremely common for people to run tunes that disable the check engine light that is produced when you delete the exhaust gas recirculation valve (EGR), urea pump and diesel particulate filter (DPF).

For things like the EGR, blocking it off is simply installing a bracket or two and closing off some vacuum lines. I can imagine people would simply reinstall this and swap the tune to pass and then immediately remove it after.

In the US, for emissions, all you have to do is not run any check engine lights when at the test center which is easy to do in the software tune.

> In the US, for emissions, all you have to do is not run any check engine lights when at the test center which is easy to do in the software tune.

Which is very disappointing...

When I first moved to Arizona I had a '67 Galaxy and that thing got to run on the dyno in the emissions office but when I got my truck to replace it (not having AC in Arizona isn't fun) all they did was plug in into the computer and charge me more for the test.

My (carburated) motorcycle was in between, they couldn't run it on the dyno and couldn't plug a computer into it so just tested it at idle -- which seemed pretty pointless to me, even being all hopped-up it still got better gas mileage than most cars on the road, though it did get pretty horrible gas mileage for a motorcycle considering I had to run premium in it.

In Spain we have a similar exam and unfortunately I know that there are indeed people that cheat. I don't know the details but there are ways to make their car emit less smoke only that day, and then go back to "normal".

Probably we should do as many Asians do in this, and make it normal to go out with a face mask.

Not sure what kind of masks have to be used to effectively not breathe pollution, and their effectivity, though. Getting informed about that is in my to-do list.

no mask can filter particles

Filtering airborne particulates is pretty much the baseline reason to use masks ¯\_(ツ)_/¯

(However surgical masks aren't meant to do that, they're only meant to catch things expelled from you, not the other way around. So I guess your interjection is somewhat correct in that running around with a surgical mask does pretty much nil to improve the quality of the air you inhale.)

Doctors also have masks to protect them from external things (for example, when they deal with patients that have tuberculosis), although I've just learned from Google that they are not technically called masks, but respirators (N95 respirators).

They are quite uncomfortable to wear, though. You can breathe just fine for extended period of time but it feels (at least to me, a layman in these things) like you get less air. I wouldn't wear it as an everyday thing.

I have no idea if there are more comfortable solutions that can handle pollution.

I'm a hobbyist woodworker and so I wear a half-mask with either a simple flat P3R filter [1] or a bigger A2P3R filter stack [2] a lot when doing that. When it's hot it is a bit annoying that sweat and moisture tends to condense in the mask. Another annoying thing is that these kind of masks tend to put pressure on the nasal bridge which gives you similar issues like a poorly adjusted pair of glasses. This is a larger problem with the filter stack because it is much heavier than the flat filters.

As for efficiency, I found a paper (DGUV IFA 0233) saying standard P3 particle filters are highly efficient even against ultrafine aerosol particles (particle diameter below 100 nm); for "larger" stuff (PM1 and larger) these are specified to remove more than 99.9 % of them.

This actually leads to P3 filters already removing a lot of smells, e.g. when you machine steel or wood, sand wood or grind steel, you don't smell it. However, gaseous stuff like solvents get through. That's what the A2 filter is for.

[1] https://asset.conrad.com/media10/isa/160267/c1/-/de/831230_A...

[2] https://assets.tooler.de/cimage/700x700/media/catalog/produc... [note: outer particle filter isn't installed in this picture, it would go under the translucent hood and is about the size of a Leibniz keks [3]

[3] https://en.wikipedia.org/wiki/Leibniz-Keks

I've worked in one and I got used to it, wearing it for an hour at a time for 4-5 hours per day. For comfort, a positive pressure air purifying respirator (PAPR) is preferable but looks like you're on a biohazard scene.

Do you have any link or description of such a mask?

" Schmied responded that as long as emissions went down when limits were tightened, his department didn’t mind they were many times higher than allowed."

Some context for this: the official European test cycle was known not to be representative of real driving conditions. Because it was less aggressive than real-world driving, it pretty much always overstated fuel efficiency and understated emissions. So real-world emissions not going down at the same rate as tested emissions is a sign of foul play, but them being higher isn't necessarily. (The US test cycle is better.)

> The US test cycle is better.

Then you never heard about the US test cycle cheating scandal. It's not better, it's actually simplier to cheat. The US test cycle is tailored to a more realistic city environment with a lot of stop's and go's, cold starts, and requires an actual professional driver. No automation. These drivers learned soo much how to cheat the system, that they were in high demand and they were all held exclusively by the US car industry, so Asian and European companies were at disadvantage. That's why the Asian industry agreed to use the European test cycles and not the more realistic US system. It was too expensive and too easily tricked. The European standard was cheaper and fairer (those times). Until they found a way to trick it also.

Sure, but don't the laws limit emissions? Irrespective of how well standard government tests were measuring those emissions?

So "We publish this data" just sounds like "I'm doing my job. Anything more is not my problem."

The laws limited emissions in the standard government test cycle, and the thresholds on emissions were chosen based on how well cars could be made to perform in those test cycles. While actual defeat devices were illegal, the legal emissions limit wasn't actually the real-world emissions limit and everyone knew it.

(The EU is now gradually moving to real-world emissions testing and a more realistic test cycle, though judging from the other reply they might have some cheating problems with that - especially as real-world testing inherently has wider errors than lab testing.)

Netflix'a Dirty Money documentary Ep 1 covers this topic: https://www.netflix.com/title/80118100

Excellent review by Adrian Colyer for those interested - https://blog.acolyer.org/2017/06/20/how-they-did-it-an-analy...

Also, the book How they did it gives insightful glimpses.

This happens in almost all industries. Big oils' repetitive attempts to kill EVs, Big Pharmas, insurance and hospitals lobbying against healthcare reforms, Big Banks working against finance regulations etc.

Big Parma is lobbying FOR healthcare reforms.

Not for anything that may reduce profits though.

So did the recall all those failing 97% of 250 car models in the EU?

Yes, but in many countries like the UK the recall is voluntary, it's just a recommendation.

Thank God it is as it worsens mileage.

Thank God it is as it allows you to place externalities on us?

I've seen claims that VW's whole diesel emissions and engine control subsystem was actually contracted out to Bosch. But VW execs knew about the software defeat device.

Afaik, yes. bosch was contracted to provide the engine control modules, including features to automatically disable the emission reduction measures. Boach apparently wrote in the contracts that such disablig was only to be used for internal testing, not in production.

The controlling unit indeed comes from Bosch, but it is a generic unit that can control any kind of engine. The car manufacturer using this unit programs it to perform the actual world, mostly by specifying parameter sets for its operation. The car manufacturer has for example to determine, how exactly the ignition needs to be set up for a given temperature and load etc. There are plenty sensor inputs these profiles can depend on like engine temperature, revs, load, outside temperature, speed of the car. The Bosch software can offer the "test dectection" as one of the inputs. There can be quite reasonable use cases for that information, but that VW used it in the way they did, was VWs choices. The same or a similar control from Bosch unit is probably used by many other car manufacturers, but without abusing this specific "feature".

As the article writes, there are other "strange" engine configurations with other car manufacturers, like using the outside temperature. Reducing or switching off the exhaust cleaning system at outside temperatures below 17C (the tests are guaranteed to be done at 20C) achieves a similar effect to what VW was doing without literally breaking the law by using a defeat device.

> There can be quite reasonable use cases for that information

What reasonable use case is there for a detection of the test cycle just from ECU information? If you're just testing stuff surely you have access to the ECU to just set a flag manually for anything you need?

Could be reasonabe safety features like not performing emergency breaking, or functions aiding the testing process (without interfering with it of course).

I can see plenty of reasons to change behavior for the test. But that's not the question. The question is why infer you are being tested through the ECU inputs? Setting a flag manually seems like a much better solution for all those objectives versus a possibly unreliable inference from speed and other data.

I could never understand how something that was less refined, diesel, meaning that it had more unnecessary “stuff” could produce less emissions. Diesel always did, and always will, as far as I’m concerned he snake oil when it comes to emissions.

Its less refined in the sense of less cracking.

It contains more energy for a given volume, so you get better fuel economy than with petrol. It might not be better for the particulates and NOx that we are now optimising for, but its better for CO2, that we were optimising for. So I don't think the snake oil label is warranted.

Diesel engines are also significantly more efficient at converting that energy into usable torque. That's why an idling diesel consumes almost no fuel - it isn't doing any work.

I always cringe when one of these "we could have been driving electric cars" articles shows up. Cars are not bread and the vast vast majority of the planet's population (who even own a car) don't just go around shopping for new cars. It's going to take DECADES before electric cars have a significant impact on the climate change. That is an economic reality. (Not to mention that electric cars are only now starting to barely match gasoline cars in terms of usability and range.)

A coworker and I had a discussion with another coworker when this scandal broke. We couldn't convince him that he was wrong when he said there is no way car software could "cheat" like that. Even after we came up with several things that could be done in the space of about a minute, he remained unconvinced. If you follow through the article to the reference US DoJ press release, VW's software used several of the ideas we came up with as part of their "cheat". This coworker also thought that the California drought was a big hoax, so...

What's the point of this anecdote other than to make you look smart and your coworker look dumb? Maybe both things are true but it's hardly a very interesting observation that there are both smarter and dumber people in the population.

It is how scandals work. Lots of people believe the lies and can't be persuaded to think from an evidence first viewpoint. They need the story to be headline news on the BBC or some other reputable source before they will 'believe it'.

Every scandal relies on groupthink to keep it quiet.

You know I wouldn't have have read his comment if you hadn't wrote this one.

in the middle he points how the software "features" were decidedly malicious as they are similar to their own hypothetical solutions to circumvent the law?

Applications are open for YC Winter 2020

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact